• 综合
  • 标题
  • 关键词
  • 摘要
  • 学者
  • 期刊-刊名
  • 期刊-ISSN
  • 会议名称
搜索

作者:

Li, Jianqiang (Li, Jianqiang.) | Cheng, Wenxiu (Cheng, Wenxiu.) | Xu, Xi (Xu, Xi.) | Zhao, Linna (Zhao, Linna.) | Liu, Suqin (Liu, Suqin.) | Gao, Zhengkai (Gao, Zhengkai.) | Ye, Caihua (Ye, Caihua.) | You, Huanling (You, Huanling.)

收录:

EI Scopus SCIE

摘要:

Airborne pollen identification is crucial to help patients prevent pollinosis symptoms. Existing data-driven methods rely on large-scale pollen images with simple backgrounds. In real scenarios, the background is complex and the data scale is small. Therefore, these methods suffer from two challenges: (1) Irrelevant information interference; (2) Incomplete feature attention. To overcome these challenges, we propose a prior knowledge-guided deep feature learning (PK-DFL) for real-world optical microscope image classification. Its main steps are as follows: Pollen location is designed to locate pollen grains based on color features, aiming to boost the accuracy of shape and texture prior feature extraction. Shape-texture awareness helps to extract the shape and texture of pollen grains via predefined feature extractors (i.e., a set of shape descriptors and an improved SFTA). These features are used to construct two types of prior knowledge, namely shape-texture attention maps (STA maps) and shape-texture feature vectors (STF vectors). Pollen classification uses a deep network (CNN) to classify pollen via imitating the pollen identification procedure of palynologists. It uses STA maps to weight pollen images and convolutional feature maps for instructing the CNN to focus on critical areas of pollen images ( for the first challenge). ). STF vectors are employed to obtain the inter-class similarity of pollen via template matching. This information is further converted to soft targets that are used to supervise the CNN attending to comprehensive key features ( for the second challenge). ). Extensive experiments on real-world datasets demonstrate the effectiveness of our PK-DFL (with accuracy and F1-score over 88%).

关键词:

Knowledge distillation Attention mechanism Optical microscope Real-world pollen classification Prior knowledge

作者机构:

  • [ 1 ] [Li, Jianqiang]Beijing Univ Technol, Sch Software Engn, Beijing 100124, Peoples R China
  • [ 2 ] [Cheng, Wenxiu]Beijing Univ Technol, Sch Software Engn, Beijing 100124, Peoples R China
  • [ 3 ] [Xu, Xi]Beijing Univ Technol, Sch Software Engn, Beijing 100124, Peoples R China
  • [ 4 ] [Zhao, Linna]Beijing Univ Technol, Sch Software Engn, Beijing 100124, Peoples R China
  • [ 5 ] [Liu, Suqin]Beijing Univ Technol, Sch Software Engn, Beijing 100124, Peoples R China
  • [ 6 ] [Gao, Zhengkai]Beijing Univ Technol, Sch Software Engn, Beijing 100124, Peoples R China
  • [ 7 ] [Ye, Caihua]Beijing Meteorol Serv Ctr, Beijing 100089, Peoples R China
  • [ 8 ] [You, Huanling]Beijing Meteorol Serv Ctr, Beijing 100089, Peoples R China

通讯作者信息:

  • [Xu, Xi]Beijing Univ Technol, Sch Software Engn, Beijing 100124, Peoples R China;;

查看成果更多字段

相关关键词:

来源 :

EXPERT SYSTEMS WITH APPLICATIONS

ISSN: 0957-4174

年份: 2024

卷: 237

8 . 5 0 0

JCR@2022

被引次数:

WoS核心集被引频次: 3

SCOPUS被引频次: 5

ESI高被引论文在榜: 0 展开所有

万方被引频次:

中文被引频次:

近30日浏览量: 1

归属院系:

在线人数/总访问数:502/4950597
地址:北京工业大学图书馆(北京市朝阳区平乐园100号 邮编:100124) 联系我们:010-67392185
版权所有:北京工业大学图书馆 站点建设与维护:北京爱琴海乐之技术有限公司